REPORT OF THE JOINT INQUIRY INTO THE TERRORIST ATTACKS OF SEPTEMBER 11, 2001

Community was not, however, poised or equipped to deliver the kind of analytic products needed. The FBI, for example, was not even aware of the collective significance of information pertaining to al-Qa'ida that was contained within its own files. This fact is underscored by its failure to connect available information on al-Mihdhar and al-Hazmi, Zacarias Moussaoui, and the FBI Phoenix field office agent's Electronic Communication in the summer of 2001. The FBI's Deputy Assistant Director for Counterterrorism Analysis, recently detailed from CIA to improve the FBI's analytic capability, testified that the Bureau "didn't have analysts dedicated to sort of looking at the big picture and trying to connect the dots, say between the Phoenix memo and Moussaoui and some [page 66] other information that might have come in that might have suggested that there were individuals there who might be preparing to hijack aircraft."

One of the primary reasons that there was so little focus on strategic analysis in the Intelligence Community may have been the perception that operational personnel and matters were more important to agency counterterrorism missions and operations than analysis and analytic personnel. Consistent with its traditional law enforcement mission, the FBI was, prior to September 11, a reactive, operationally driven organization that did not value strategic analysis. While FBI personnel appreciated case specific analysis, for example, most viewed strategic analytic products as academic and of little use in ongoing operations. The FBI's Assistant Director for Counterterrorism acknowledged in Joint Inquiry testimony that the reactive nature of the FBI was not conducive to success in counterterrorism:

No one was thinking about the counterterrorism program what the threat was and what we were trying to do about it. And when that light came on, I realized that, hey, we are a reactive bunch of people, and reactive will never get us to a prevention and what we do. . . .Is there anybody thinking and where's al-Qa'ida's next target? And no one was really looking at that.

He also testified about the difficulty of going beyond the FBI's traditional case-oriented approach:

We will never move away from being reactive. We understand that. And that's what people want to talk about most of the time is how's that case going in East Africa, or how's the USS Cole investigation going? But if you step back and look at it strategically you need to have people thinking

62

beyond the horizon and that's very difficult for all of us. It's particularly difficult for law enforcement people.

Other FBI executives acknowledged the FBI's pre-September 11 analytic failings. Director Mueller testified that:

I would be the first to concede that we have not done a good job in analysis. We have not had either the technology nor the analytical cadre of individuals that we have needed to perform strategic analysis.

In Joint Inquiry testimony, the FBI's Deputy Assistant Director for Counterterrorism Analysis referred to strategic analysis as the FBI's "poor stepchild" prior to September 11, 2001. As a result, our review confirmed that strategic analysts were often marginalized by the operational units and rarely, if ever, received requests from operational sections for analytical assessments of pending al- Qa'ida's cases.

In 2000, FBI management aggravated this situation by transferring five strategic analysts who had been working on al-Qa'ida matters to FBI operational units to assist with ongoing cases. According to a former Chief of the International Terrorism Analytic Unit, this "gutted" the analytic unit's al- Qa'ida-related expertise and left the unit with little ability to perform strategic analysis.

Concerns about protecting criminal prosecutions also limited the FBI's ability to utilize strategic analytic products. In interviews, some analysts said they frequently were told not to produce written analyses, lest the analyses be included in discovery during criminal prosecutions. FBI analysts were further hindered because of the limitations of the FBI's information technology.

Due in large part to these cultural and practical issues, the Bureau has had little success in building a strategic analytic capability, despite numerous attempts before September 11 to do so. For example, in 1996, the FBI hired approximately fifty strategic analysts for counterterrorism purposes, many with advanced degrees. According to both current and former FBI analytic personnel and supervisors, most of those analysts left the Bureau within two years because they were dissatisfied with the role of strategic analysis at the FBI.

63

The lack of emphasis on strategic counterterrorism analysis was also an issue at the CIA. The former Chief of CTC testified that, at the CTC:

We have under-invested in the strategic only because we've had such near-term threats. The trend is always toward the tactical. . . . The tactical is where lives are saved. [Page 68] And it is not necessarily commonly accepted, but strategic analysis does not . . . get you to saving lives.

Analysts in the CTC also expressed concern to the Joint Inquiry that their opinions were not given sufficient weight. A manager in the CTC confirmed to the Staff that CIA operations officers in the field resented being tasked by analysts because they did not like "to take direction from the ladies from the Directorate of Intelligence." Despite the need for increased analytic capability, CTC reportedly refused to accept analytic support offered by at least two other agencies prior to September 11, 2001. As mentioned earlier, representatives of both FAA and DIA informed the Inquiry that CTC management rebuffed their offers of analytic assistance in 2000 because those agencies wanted greater access to CTC information in return, and this raised CTC concerns regarding protection of its intelligence sources and methods.

Analysts at NSA commented to the Joint Inquiry that CTC viewed them as subordinate - "like an ATM for signals intelligence." NSA analysts say they attempted to accommodate CTC preferences by focusing on short-term operational requirements - sometimes at the expense of more thorough analysis -- and even altered NSA reporting formats because CTC did not like including NSA analyst comments in the text of signals intelligence reports. Several NSA analysts also described a definite perception that the DCI would always side with CIA and CTC operational personnel in any disagreements between NSA and CTC.

Some of the shortcomings in analytical capability can be traced to the fact that analysts were often inexperienced, under-trained, and, in some cases, unqualified for the responsibilities they were given. At the CTC, the analysts were a relatively junior group prior to September 11 since CTC had traditionally relied on rotational assignments. An analytic career service was not created in CTC until about 1997. The average CTC

64

analyst had three years of analytic experience, versus the eight years for analysts in the CIA's Directorate of Intelligence.

[Page 69]

A former counterterrorism analyst at DIA explained to the Joint Inquiry the consequences for analytic perspective of this shortfall in experience and knowledge:

Coupled with this issue of experience comes the ability to place current intelligence reporting in the context of historical perspectives. In the period leading up to the 1998 East Africa bombings and the 2000 attack against the USS Cole in Yemen, terrorism analysts nearly across the board incorrectly assessed that a group would not conduct an attack in an area where it was able to operate with relative ease. Additionally, there appears to be a continued reluctance to correctly assess and evaluate the nature of cooperation between many [ ] and [ ] Islamic extremist groups. Both of these examples, and there are certainly others, occurred despite over a decade of credible reporting to the contrary.

At the FBI, a January 2002 internal study found that 66% of the FBI's 1200 "Intelligence Research Specialists" (strategic analysts) were unqualified. This problem was compounded by the fact that newly-assigned strategic and operational analysts received little counterterrorism training upon assuming their positions. As the Chief of the FBI's National Security Intelligence Section testified:

While there was no standardized training regimen, other than a two-week basic analytical course, training was available on an ad hoc basis and guidance was provided by both the unit chiefs of the analytical units and the FBI's Administrative Services Division. The development of a standardized curriculum, linked to job skills, and career advancement was being planned . . . , but it was never implemented.

The quality of Intelligence Community counterterrorism analysis also suffered as a result of the fact that agency analysts often did not have access to important information residing at other agencies. DIA's Associate Director for Intelligence at the Joint Chiefs of Staff testified about the extent of these problems:

In my opinion, one of the most prolonged and troubling trends in the Intelligence Community is the degree to which analysts, while being expected to incorporate all sources of information into their assessments,

65

have been systematically separated from the raw material of their trade…. At least for a few highly complex high stakes issues, such as terrorism, where information by its nature is fragmentary, ambiguous and episodic, we need to find ways to emphatically put the "all" back in the discipline of all-source analysis.

[Page 70]

Intelligence Community analysts had particularly limited access to "raw material" contained in the FBI's counterterrorism investigations, including Foreign Intelligence Surveillance Act (FISA)- erived information, and to unpublished NSA information. The former acting head of the FBI's Usama Bin Ladin Unit informed the Joint Inquiry that, prior to September 11, the FBI would generally only provide the CIA with FISA-derived information when the FBI wanted it passed to a foreign government. Primarily due to the FBI's technological problems, the FBI's analysts did not even have access to all relevant FBI information. The FBI's Deputy Assistant Director for Counterterrorism Analysis testified that "the FBI lacked effective data mining capabilities and analytical tools, it has often been unable to retrieve key information and analyze it in a timely manner-and a lot has probably slipped through the cracks as a result."

There also was, and apparently continues to be, a reluctance at CIA to provide raw data to analysts outside the Agency. DCI Tenet testified that even analysts at the Department of Homeland Security will not be allowed access to CIA raw data:

There was a headline today that said raw data provided. Well, actually that's not what's envisioned. They will get all of the finished product, the finished analytical product, the finished intelligence that NSA, CIA and FBI issues, and on a case-by- case basis, depending on what kind of an environment we're in, we actually may give them a piece of raw data.

Discussions of access to "raw data" or "raw traffic" raise objections from CIA, since it immediately equates the term to internal operational traffic, and from NSA. Both agencies are concerned with protecting the sources and methods they use to collect intelligence, a responsibility that has been specifically placed upon the DCI by the National Security Act, and NSA is also concerned about its legal responsibilities to "minimize" U.S. person data in the information it collects.

66

A significant portion of the communications collected by NSA involves U.S. persons as parties or contains information about U.S. persons. NSA is responsible under law and Attorney General procedures for ensuring that information of this type that does [page 71] not have intelligence value is eliminated before intelligence is disseminated to persons outside the NSA production chain. NSA does allow analysts from other agencies to have access to raw intercepts on a case-by-case basis, typically at NSA and after the analysts have been trained in the minimization rules.

Analysts, for their part, maintain that there is intelligence information of potential significance embedded in the raw CIA and NSA data. Much of this, they believe, is filtered out during the CIA and NSA processes that determine what information analysts receive in disseminated form. The CIA has implicitly recognized this by integrating its counterterrorism analysts into CTC where they have full access to raw traffic, an access that most CIA analysts do not routinely enjoy.

[As an example, the Joint Inquiry found numerous operational cables relating to the meeting in Malaysia that was attended by al-Hazmi and al-Mihdhar in January 2000 containing information that could have enabled all-source analysts to assess that meeting more completely. DIA identified four specific leads its terrorism analysts could have pursued had this information been shared with it in early 2000, and three leads in the critical August 2001 timeframe that DIA believes would have allowed additional action to be taken concerning the arrest of Moussaoui and the watchlisting of Al-Mihdhar and al-Hazmi. However, DIA did not learn of this operational traffic until informed of it in the course of the Joint Inquiry in April 2002].

Intelligence analytical personnel told the Joint Inquiry that they are not seeking access to operational details or the identification of sources and methods. The DIA Director, for example, observed that he has tried to convince CTC that DIA does not want operational details, but only important intelligence buried in the operational traffic.

The inadequate quality of the Intelligence Community counterterrorism analysis impacted not only the Intelligence Community's strategy and operations, but also the

67

ability of the U.S. Government's policymakers to understand the threat and to make informed decisions. Several current and former U.S. government policymakers provided [page 72] testimony to this effect before these Committees. For example, Richard Clarke, the former National Coordinator for Security, Infrastructure and Counterterrorism at the National Security Council ("National Counterterrorism Coordinator") explained to the Joint Inquiry that:

FBI did not provide analysis. FBI, as far as I could tell, didn't have an analytical shop. They never provided analysis to us, even when we asked for it, and I don't think that throughout that 10-year period we really had an analytical capability of what was going on in this country.

Richard Armitage, the Deputy Secretary of State complained that Intelligence Community analysis tends to provide policymakers with only one view, and that dissenting opinions are rarely expressed:

I am the consumer. It's very rare that we get the one off voice or the dissident voice . . . . For a policy maker, the dissident voice is very helpful to either confirm what you think or really open up a new area, and this is not generally done. If I had to say the one biggest weakness in the analysis area, I would say that's it. Second, it's the way analysis in the Intelligence Community is generally put forth, and it's related, and that is consensus…I really would just enforce this observation about the need to get alternative views up, because almost everything that's important here is shrouded in ambiguity and uncertainty. There is a tendency to want to get things scrubbed out to get the differences eliminated.

Former National Security Advisor Sandy Berger implied in his testimony that the U.S. Government has often relied too heavily on analytic expertise within the U.S. Government, and that he believes that the best analytic expertise is often found elsewhere:

I think we live in a world . . .in which expertise increasingly does not exist in the government. It's a very complicated world. And the five people who know Afghanistan the best or Sierra Leone the best are probably located either in academia, think tanks or in companies, not to devalue the people of the government. So we have to find a way in my judgment to integrate the expertise that exists on the outside with the information that exists on the inside.

A former DIA counterterrorism analyst told the Joint Inquiry hearing on October 8, 2002: [page 73]

68

The single most important issue that will affect future performance is the experience level of the analyst. While this certainly applies to all intelligence analysts regardless of subject area, it is even more critical for those trying to prevent the next terrorist attack. In the case of an analyst responsible for tracking a Middle Eastern terrorist group, this person will need to have an expertise or at least a good working knowledge of terrorism itself, the group that they have for an account, regional and country issues present in the group's operating area, which can be quite extensive, and Islamic history, culture and the sects thereof. This . . . required level of expertise is rarely going to be found outside the Intelligence Community and is instead going to be recruited from academia and then developed in-house through training programs and mentors.

Former Chairman of the House Intelligence Committee Lee Hamilton noted in his testimony to the Joint Inquiry on October 3, 2002 that the Hart-Rudman Commission had concluded that the U.S. Government's personnel system has become a national security issue. As he stated:

There is too much rigidity in the system. There is not enough allowance for incentive. And it is an exceedingly serious problem in our government. And it has national security consequences. We've got to work through this matter so that managers can manage more effectively. . . . . I would absolutely assure you . . . that you would not tolerate in your office the kind of management restrictions that operate today in the federal government. . . . Now I know the importance of this to employees, so it's a tough problem, but the only thing I want to say here, Senator, when you talk about personnel we are now approaching this national security review and we have to look at the civil service system and we have to find ways and means of getting more flexibility into it. If we don't, we're going to choke ourselves to death.

During the same hearing, former CIA Inspector General Frederick Hitz discussed a number of actions that might be taken to enhance the quality of the personnel employed by the Intelligence Community agencies. These included the idea of establishing an intelligence reserve corps that could be activated at a time of particular need, an intelligence reserve officer training corps, and more internships to introduce young people into the agencies. While he recognized that some of these ideas are not new, he did not believe they had been vigorously pursued.

In sum, prior to September 11, the Intelligence Community's analytic components failed to understand the collective significance of the information in their possession.

69

This failure is attributable not only to the factors discussed above, but also to a basic lack of creativity and imagination in evaluating the intelligence that was at hand. Ironically, the best example of the creative, imaginative and aggressive analysis of relevant [page 74] intelligence that this review has found was not a product of Intelligence Community analysts, but, instead, of an FBI field agent in Phoenix. The Phoenix agent, in reviewing his office's case files, went beyond the facts of those individual cases to focus on a larger, and far more serious, picture of the potential, long-term threat. By putting together various pieces of information, he became convinced that Usama Bin Ladin was sending individuals to aviation-related training in order to put al-Qa'ida in a position to target civil aviation. His July 2001 Electronic Communication to FBI Headquarters was a strategic analytic product that correctly identified at least one critical element that was to be used in the plot that unfolded on September 11, an element that apparently eluded far more seasoned analysts elsewhere in the Intelligence Community.

6. Finding: Prior to September 11, The Intelligence Community was not prepared to handle the challenge it faced in translating the volumes of foreign language counterterrorism intelligence it collected. Agencies within the Intelligence Community experienced backlogs in material awaiting translation, a shortage of language specialists and language-qualified field officers, and a readiness level of only 30% in the most critical terrorism-related languages.

Discussion: The language problem has been one of the Intelligence Community's perennial shortfalls. Prior to September 11, the shortage of language specialists who would be qualified to process large amounts of foreign language data in general, and Arabic in particular, was one of the most serious issues limiting the Intelligence Community's ability to analyze, discern, and report on terrorist activities in a timely fashion. According to a senior NSA official, [ ]. These are promptly scanned for intelligence value, and only the most important - [ ] -- are then translated into English. Yet, prior to September 11, NSA had [ ] personnel assigned to this task.

[Analyzing, processing, translating, and reporting al-Qa'ida-related [ ] communications requires the highest levels of language and target knowledge expertise that exist at the National Security Agency. The large number of

70

communicants whose native origins cover all of the major Arabic dialects makes this [page 75] analysis linguistically and analytically difficult. The target lives in and understands life in a thoroughly Islamic milieu, a milieu that is often reflected in the target's communications].

Evaluating these communications requires considerable subject matter expertise in Islam in general and Islamic extremism in particular in order to ensure the best possible interpretations. Very few Arabic language analysts at NSA have done any graduate work in Islamic Studies and the vast majority of these linguists [].

The NSA Senior Language Authority explained to the Joint Inquiry that the Language Readiness Index for NSA language personnel working in the counterterrorism "campaign languages" is currently around 30%. This Index is based on the percentage of the mission that is being performed by qualified language analysts. The current low level of the Index is due in part to the fact that NSA has moved roughly [ ] language personnel since September 11 from areas in which they were performing quite well to counterterrorism, where they must gain experience and expertise before their performance can improve.

[According to the Chief of the FBI's Language Services Division, prior to September 11, the Bureau employed [ ] Arabic speakers and was experiencing a translation backlog. As a result, 35% of Arabic language materials derived from Foreign Intelligence Surveillance Act (FISA) collection were not reviewed or translated. If the number of Arabic speakers were to remain at [ ], the projected backlog would rise to 41% in 2003.]

The Director of the CIA Language School testified that, given the CIA's language requirements, the CIA Directorate of Operations is not fully prepared to fight a world- [page 76] wide war on terrorism and at the same time carry out its traditional agent recruitment and intelligence collection mission. She also added that there is no strategic

71

plan in place with regard to linguistic skills at the Agency. When asked about the language ability of CIA field officers, the Language School Director stated:

[Traditionally we have had an adequate number of Arabic speakers to conduct their business in [ ]. Level of language required to use with a volunteer or for a thorough debriefing is very different than the level of language you need to socially chit-chat with somebody or to even recruit someone. And that is where the bar has been raised much higher, and that's why we must now have a cadre of language speakers, [ ] who indeed can debrief and write up reports with these volunteers].

The Director of the CIA Language School explained that CIA should have a pool of interpreters to meet language support needs at home and abroad, but that this is not easy to achieve. She stated that: "With the progress of technology, we keep on getting more material - [ ]. These things need translation, we don't have that capability." In her view, CIA field officers are typically generalists, and this has been important to their career progression culture since the mid- 1970s. Now, however, it is an absolute must that these officers possess expertise rather than mastery of "one little dab here and one little dab there." Her recommendation was that either a culture change within CIA is called for or that a cadre of specialists be developed and not penalized.

7. Finding: [Prior to September 11, the Intelligence Community's ability to produce significant and timely signals intelligence on counterterrorism was limited by NSA's failure to address modern communications technology aggressively, continuing conflict between Intelligence Community agencies, NSA's cautious approach to any collection of intelligence relating to activities in the United States, and insufficient collaboration between NSA and the FBI regarding the potential for terrorist attacks within the United States].

Discussion: While one of the Intelligence Community's greatest strengths is its ability to rely on its advanced technical collection capabilities, the Joint Inquiry confirmed that the Community did not, prior to September 11, fully exploit those [page 77] capabilities in the effort against Bin Ladin and al-Qa'ida. Pre-September 11, [ ]. Post-September 11, this increased to [ ].

72

It became very clear after September 11 [ ]. In testimony before the Joint Inquiry, the NSA Director acknowledged that "little was known prior to 11 September of how al-Qa'ida used [ ] communications. . . .We continue to attack key gaps that remain in our . . . [ ] exploitation capabilities."

Similarly, NSA has long had a program to use [ ], but again little was known about al- Qa'ida targets and few such operations were mounted before September 11. After September 11, this changed and the NSA Director was able to testify that: "[ ]."

The inability to bring technical collection capabilities to bear in the counterterrorism area was particularly apparent in regard to signals intelligence that could have shed greater light on the potential for terrorist activity within the domestic United States. Both the NSA and the FBI have the authority, in certain circumstances, to intercept international communications, to include communications that have one communicant in the United States and one in a foreign country, for foreign intelligence purposes. While those authorities were intended to insure a seamless transition between U.S. foreign and domestic intelligence capabilities, significant gaps between those two spheres of intelligence coverage persisted and impeded domestic counterterrorist efforts.

[Page 78]

Before September 11, it was NSA policy not to target terrorists in the United States, even though it could have obtained a Foreign Intelligence Surveillance Court order authorizing such collection. NSA Director Hayden testified that it was more appropriate for the FBI to conduct such surveillance because NSA does not want to be

73

perceived as targeting individuals in this country and because the intelligence produced about communicants in the United States is likely to be about their domestic activities.

[As a result, NSA regularly provided information about these targets to the FBI - both in its regular reporting and in response to specific requests from the FBI - [ ] that NSA acquired in the course of its collection operations. The FBI used this information in its investigations and obtained FISA Court authorization for electronic surveillance [ ] when FBI officials determined that such surveillance was necessary to assist one of its intelligence or law enforcement investigations].

[One collection capability that was used by both NSA and FBI under approval of the FISA Court (the "FISA Court technique") had a [ ] probability of collecting [ ] communications between individuals in the United States and foreign countries. NSA did not use the FISA Court technique against [ ], however, precisely because of this [ ] probability].

As NSA Director Hayden has testified to the Joint Inquiry, NSA believed it was the FBI's responsibility to collect communications of individuals in the United States. General Hayden stated two reasons for this position. One is that, since the individual is in the United States, the information obtained is most likely to relate to domestic activity that is of primary interest to the FBI. The second reason is that NSA does not want to be viewed as targeting persons in the United States. Joint Inquiry interviews of a wide range of NSA personnel, from the Director down to analysts, revealed the consistent theme that NSA did not target individuals in the United States. This is so ingrained at NSA that one counterterrorism supervisor at NSA admitted that she had never even thought about using this technique against [ ].

[Page 79]

Despite the NSA view that this category of intelligence collection was the FBI's responsibility, NSA and the FBI did not develop any plan to ensure that the Bureau made an informed decision about whether to use the FISA Court technique to collect communications between the United States and foreign countries that NSA was not covering. Thus, a gap developed between the level of coverage of communications

74

between the United States and foreign countries that was technically and legally available to the Intelligence Community and the actual use of that surveillance capability].

[This gap was potentially very damaging in the case of Khalid al-Mihdhar during the period in early 2000 when he was in the United States. [ ]. His presence in the United States was not determined by the Intelligence Community at the time. [ ].

[NSA and CIA officers often worked closely together in [ ] collection efforts against al-Qa'ida. The two agencies conducted [ ] operations, And these operations often met with some success. However, one type of these operations - [ ] - caused much friction between NSA and CIA. This was especially true at the mid- and upper-management levels where struggles developed regarding which agency was in charge of developing and using such technology when human intelligence and signals intelligence targets overlapped. CIA perceived NSA as wanting to control technology deployment and development, while [page 80] NSA was concerned that CIA was conducting NSA- type operations. The NSA Chief of Data Acquisition noted to the Inquiry that this has been an issue during his entire tour of almost three years. These frictions persisted even after the September 11 attacks. In the first six months of 2002, for example, no less than seven executive-level memoranda (including one from the President) were issued in attempts to delineate CIA and NSA responsibilities and authorities in this collection area].

The Chief of NSA's Signals Intelligence Directorate acknowledged these frictions in a Joint Inquiry interview, but cited the executive memoranda as evidence that the situation is improving. NSA Director Hayden, told the Joint Inquiry that "the old divisions of labor are impractical; the new electronic universe requires more and more

75

cooperation. " He also added that he "would not be surprised if someday the closeness of this relationship would require organizational changes."

8. Finding: The continuing erosion of NSA's program management expertise and experience has hindered its contribution to the fight against terrorism. NSA continues to have mixed results in providing timely technical solutions to modern intelligence collection, analysis, and information sharing problems.

Discussion: One of the side effects of NSA's downsizing, outsourcing, and transformation has been the loss of critical program management expertise, systems engineering, and requirements definition skills. These skills were devalued by NSA during the 1990s when most technical development was done within the agency, and the impact of their loss was evident in NSA's response to the Joint Inquiry's attempts to gather information concerning NSA's plans for developing solutions to its current technology gaps in areas of particular importance to counterterrorism. [ ]. NSA was able to provide little more than very high-level and general vision statements.

The impact of this lack of program management was evident during interviews with analysts who expressed frustration regarding their current working environment. [page 81] For example, they must now write three versions of reports in order to accommodate the demands of various customers and uses. The TRAILBLAZER program, which the NSA Director has described as NSA's "effort to revolutionize how we produce SIGINT in a digital age," is now not expected to produce such results until 2004 at the earliest and confusion still exists as to what those results will actually be. In the meantime, none of the analysts were aware of any near term efforts to alleviate their current system's technical limitations.

NSA personnel also stated that NSA's efforts to collect [ ], reveals a critical deficiency in its capabilities. The solution to this deficiency is well understood and estimated to cost less than $1 million to implement. However, the project manager is still struggling for funds to pay for an upgrade that would not be completed until 2004.

76

The Joint Inquiry also found a high level of frustration among contractors who do business with the NSA. Common themes repeated to the Joint Inquiry concern the extremely poor quality of solicitation packages and acquisition expertise on the part of NSA employees and the inability of program managers to speak with consistency and authority on future contract opportunities. NSA also lacks a formal Contracting Officer Technical Representative certification program. This is of special concern as NSA continues to increase its reliance on contractors. In testimony to the Joint Inquiry in October 2002, the NSA Director stated that NSA "spent about a third of our SIGINT development money this year making things ourselves. Next year the number will be [dropping to] 17%."

The Chief of Staff for NSA's Signals Intelligence Directorate (SID) told the Joint Inquiry he fears that "SID has lost its business acumen…and [he] worries greatly about the lack of acquisition experience and program planning, especially in light of NSA's huge budget increase." He also told the Joint Inquiry that he has worked actively on this issue, especially in providing program management training to frontline workers.

[Page 82]

9. Finding: The U.S. Government does not presently bring together in one place all terrorism- related information from all sources. While CTC does manage overseas operations and has access to most Intelligence Community information, it does not collect terrorism-related information from all sources, domestic and foreign. Within the Intelligence Community, agencies did not adequately share relevant counterterrorism information, prior to September 11. This breakdown in communications was the result of a number of factors, including differences in the agencies' missions, legal authorities and cultures. Information was not sufficiently shared, not only between different Intelligence Community agencies, but also within individual agencies, and between the intelligence and the law enforcement agencies.

Discussion: Counterterrorism, like other transnational threats such as drug trafficking, requires close coordination and information sharing among and within the Intelligence Community agencies. Despite some improvement, significant problems remained in the sharing of information within the Intelligence Community, prior to September 11. As a result, the Community was unable to exploit the full range of capabilities and expertise in the counterterrorist effort.

77

Each of the principal collectors and analyzers of counterterrorism intelligence -- the FBI, CIA, NSA, and DIA -- has its own distinct missions, sets of legal authorities and restraints, and cultures. Unfortunately, these factors, while serving many other legitimate purposes, often hinder collaboration and willingness to share information. In his testimony, former Congressman and House Intelligence Committee Chairman Lee Hamilton described the problem:

The very phrase "Intelligence Community" is intriguing. It demonstrates how decentralized and fragmented our intelligence capabilities are. . . . The Intelligence Community is a very loose confederation. There is a redundancy of effort, an imbalance between collection and analysis, and problems, as we have repeatedly heard in recent weeks, of coordination and sharing.

While DCI George Tenet and former FBI Director Louis Freeh testified that collaboration and information sharing in the Intelligence Community have markedly improved in recent years, this Inquiry found that the agencies still act too often and at too many levels as a loose collection of entities. The Joint Inquiry heard testimony that confirmed problems in sharing information between different Intelligence Community agencies, within individual Intelligence Community agencies, and between law [page 83] enforcement and intelligence agencies.

For example, the former FBI agent who had handled the San Diego informant testified about his personal experience with information sharing between the FBI and the CIA:

Ms. Hill: You also [said] that, in your opinion, information sharing between the FBI and the CIA prior to 9/11 was almost nonexistent.

Former FBI Agent: It was bad. Well, it's not nonexistent, but . . . if you have a case that has a common mission and everybody can benefit from it, you're going to get their assistance. But if you don't have that, asking them for something, it's very, very difficult.

. . . .

Former FBI Agent: If I had to rate it on a ten-point scale, I'd give them about a 2 or a 1.5 in terms of sharing information.

Ms. Hill: Well, could you tell us what your experience was? Why do you say that?

Former FBI Agent: [P]art of the problem here, I think, is being able to communicate with them. . . .

78

Ms. Hill: By "them," you mean the CIA?

Former FBI Agent: With the CIA. Everything's got to go through headquarters, usually.

Ms. Hill: Through your headquarters, or through CIA?

Former FBI Agent: Through [FBI] headquarters. Normally, . . . you have some information you want the Agency to check on. You end up writing it up, sending it back through electronic communication or teletype, . . . or memo. . . . And then the Bureau, FBI headquarters, runs it across the street to the Agency. And then, maybe six months, eight months, a year later, you might get some sort of response.

Even after the first World Trade Center attack in 1993, the Millennium plot, and attacks against U.S. embassies in East Africa in 1998 revealed that global Islamic extremists were capable of reaching into the United States, there was little sustained effort by the FBI, NSA, and CIA to work together to collect and share information about contacts between foreign persons in the United States and those abroad. For example, while a great amount of information that NSA collects is routinely transmitted electronically into CTC databases at CIA, this is not true of terrorist information collected domestically by the FBI.

The Acting Chief of the FBI's Radical Fundamentalist Unit, told the Joint Inquiry in an interview that, prior to September 11, the FBI would primarily think to provide the [Page 84] CIA with information obtained through FISA surveillance only when it was also being passed to a foreign government. The FBI did not share such information with CTC on a routine basis, partly due to the FBI's inadequate information technology, but also because they believed that sharing information with intelligence agencies raised legal concerns relating to the traditional separation between law enforcement and intelligence operations. As a consequence, gaps occurred in the collection and analysis of information about individuals and groups operating in the United States and abroad.

The FBI has traditionally viewed intelligence primarily as a tool for developing evidence to be used in FBI cases, rather than as the basis for valuable strategic analysis for the FBI or other intelligence agencies. As Director Mueller noted to the Joint Inquiry:

79

. . . one of the things that we have to do, and I think is changing since September 11, is for agents who are very good in the criminal sphere to look at a piece of information and not run it through the sifting that you do to determine whether it would be admissible in court. In other words, is it hearsay? Well, I am going to thrust it aside. Do I have lack of foundation? Therefore, I am going to disregard that.

Prior to September 11, FBI personnel were not trained or equipped to share intelligence developed during FBI counterterrorism investigations with the Intelligence Community or even with other units within the FBI on a regular basis. For example, after receiving the Electronic Communication from the Phoenix field office in July 2001 indicating that al-Qa'ida might be sending operatives to the United States for flight training, a Headquarters Intelligence Operations Specialist (IOS) did not send it to the FBI's analytic unit or to the CIA. Instead, the IOS forwarded it to the FBI field office in Portland, Oregon, primarily because of possible connections to an individual case there.

The Joint Inquiry's review of a July 2002 CIA cable that it found within a local FBI field office's investigative files provides another example of information sharing problems within the FBI. A CIA officer assigned to a Joint Terrorism Task Force in California sent a cable to CIA Headquarters after analyzing information gleaned primarily from a review of the local FBI field office's investigative files. He also [Page 85] provided a copy to the local FBI agent who was responsible for those files. The cable sets forth the CIA officer's concerns regarding indications that persons associated with a foreign government may have provided financial support to some of the September 11 hijackers while they were living in the United States. Those indications, addressed in greater detail elsewhere in this report, obviously raise issues with serious national implications. Nevertheless, the FBI agent to whom he provided a copy viewed it only in relation to ongoing investigations and did not consider its possible value for other cases or the FBI's national counterterrorism strategy. Thus, the FBI agent placed the cable in only one case file and did not forward a copy to FBI Headquarters.

Similarly, the FBI typically used information obtained through the Foreign Intelligence Surveillance Act (FISA) only in connection with the cases in which it was

80

obtained and would not routinely disseminate it within the FBI or to other members of the Intelligence Community. FBI personnel advised the Joint Inquiry that FISA information was not included in the FBI's Automated Case System (ACS) because both criminal and intelligence agents had access to that system.

Culture and policy issues also limited the extent to which CIA shared counterterrorism information within the Intelligence Community. As noted earlier, a lack of focus on the domestic terrorist threat, which was viewed as an FBI, rather than CIA, mission, accounted for some information sharing problems. For example, the DCI acknowledged in his testimony that CIA was not sufficiently focused on advising the State Department to watchlist all terrorist operatives who might be traveling to the United States, even though this would provide valuable information to domestic agencies in targeting these persons at ports of entry. On at least three occasions between January 2000 and August 2001, there were opportunities to watchlist future hijackers Nawaf al- Hazmi and Khalid al-Mihdhar, but the CIA failed to do so. In his testimony on October 17, 2002, the DCI admitted this failure, attributing it to:

. . .uneven practices, bad training and a lack of redundancy. The fact that [CTC personnel] were swamped does not mitigate the fact that we didn't overcome that [with] a separate unit or better training for those people.

[Page 86]

Aside from the formal watchlist procedure, the record strongly suggests that, despite numerous related contacts with the FBI during the period, no one at CIA advised the FBI about al-Mihdhar's U.S. visa and the fact that al-Hazmi had traveled to the United States. Ironically, this occurred despite the fact that both CIA and FBI personnel were at the time working in CTC where the information was received. The CIA employee who briefed FBI personnel about al-Mihdhar on January 6, 2000, but did not mention any information about al-Mihdhar's visa and potential travel to the United States, indicated in an e-mail to a colleague at CIA that same day: "In case FBI starts to complain later . . . below is exactly what I briefed them on." This CIA employee told the Joint Inquiry that he had, at the time, been assigned to work at the FBI Strategic Information Operations Center specifically to fix problems "in communicating between the CIA and the FBI." Obviously, such problems remained.

81

The Joint Inquiry also heard from many different agencies within the Intelligence Community, most notably the Defense Intelligence Agency (DIA), that the perception that collecting agencies have "ownership" of the intelligence they acquire impedes the free flow of information. In a Joint Inquiry interview, one DIA official complained that analysts were often denied access to critical intelligence held in other Intelligence Community agencies:

We have to get raw data to the analysts. The analysts have been separated from source-generated data that is important. There is excessive, filtering, packaging and selective product reporting that is not helpful. Some problems are so important that the U.S. Government cannot afford any longer to filter.

In his testimony, the DCI confirmed that this filtering will continue when he noted that even all- source analysts within the new Department of Homeland Security will not have access to all raw intelligence on anything like a routine basis. This tendency to ownership, in its simplest form, means that the originating agency is free to edit and otherwise truncate the information it collects before it disseminates it to other agencies. On the other hand, analysts frequently argued that, in the world of counterterrorism, there is information in this filtered data that the collecting agency may not recognize as having significance in the aggregate to analysts elsewhere. In interviews, DIA officials [page 87] emphasized that they always received threat information from other Intelligence Community agencies, but did not always have access to the background information necessary to understand the nature of the threat reporting fully. A senior DIA analytical official testified that:

Senior [Defense Department] officials received information that his analysts did not receive. However, to extract meaning from that data, to perform the true analytic function, we need to get that information into the hands and the brains of analysts who are paid to fill in the gaps of missing information to compensate for absent evidence and to turn information into knowledge. That's what we pay them to do. They don't have the information, they can't do that.

In a written statement to the Joint Inquiry, the new Director of the DIA noted: " In my opinion, one of the most prolonged and troubling trends in the Intelligence Community is the degree to which analysts - while being expected to incorporate the full

82

range of source information into their assessments - have been systematically separated from the raw material of their trade. "

Information sharing is also limited by the longstanding Intelligence Community practices of narrowly limiting disclosures of intelligence information outside normal channels in order to protect sources and methods. Disclosures to criminal investigators and prosecutors were intentionally limited to avoid having intelligence become entangled in criminal prosecutions. In deference to those kinds of restrictions, CIA did not provide the FBI New York field office criminal agents who were investigating the USS Cole bombing information regarding the al-Qa'ida meeting in Malaysia that was attended by hijackers-to-be al-Hazmi and al-Mihdhar.

A 1995 Department of Justice policy that established procedures -- often referred to as the "wall" -- governing FBI sharing of Foreign Intelligence Surveillance Act (FISA)-derived intelligence information with investigators handling parallel criminal investigations also prevented sharing of important intelligence. Under this policy, the FBI could share information from FISA surveillances with criminal investigators if the information was relevant to a crime under investigation and an attorney in an FBI field office or in the Office of Intelligence Policy and Review (OIPR) at the Department of [page 88] Justice authorized its release. In al-Qa'ida FISA cases, the FISA Court directed that the Court itself act as the "wall" and determine whether the information in question was relevant to a criminal investigation and, thus, could be shared.

Unfortunately, the Inquiry confirmed that the Intelligence Community agencies, perhaps overly "risk averse" in dealing with FISA-related matters, restricted the use of information far beyond what was required. The majority of FBI personnel interviewed in the course of the Inquiry incorrectly believed that the FBI could not share FISA-derived information with criminal investigators at all or that an impossibly high standard had to be met before the information could be shared. Most did not know that FISA-derived information could be shared with criminal investigators if it was simply relevant to the criminal investigation. Because of these misunderstandings, FBI intelligence investigators rarely sought approval to pass FISA-derived information to FBI criminal investigators.

83

Further, as a result of the FISA Court decision, NSA placed a caveat on all its [ ] terrorism intelligence products requiring OIPR approval before information could be shared with criminal investigators. This stemmed from NSA's concern that it could not determine which of its intelligence reports were the result of information obtained through FBI-conducted FISA surveillances (and therefore subject to the "wall" requirements) and which were not. The effect of this NSA effort to comply with the FISA Court's decision was an unnecessary restriction on the sharing of NSA-acquired intelligence information with criminal investigators.

In August 2001, when the FBI was attempting to locate al-Hazmi and al-Mihdhar in the United States, an FBI Headquarters e-mail prohibited New York field office criminal agents from participating in the search because the information had originated in intelligence channels. However, because this information was not derived from a FISA surveillance, there was no reason it could not be shared with FBI criminal agents. Expressing his utter frustration with the system, a New York FBI agent responded by email:

[page 89]

Whatever has happened to this - someday someone will die - and wall or not - the public will not understand why we were not more effective and throwing every resource we had at certain "problems." Let's hope the [FBI's] National Security Law Unit will stand behind their decisions then, especially since the biggest threat to us now, UBL, is getting the most "protection."

10. Finding: Serious problems in information sharing also persisted, prior to September 11, between the Intelligence Community and relevant non-Intelligence Community agencies. This included other federal agencies as well as state and local authorities. This lack of communication and collaboration deprived those other entities, as well as the Intelligence Community, of access to potentially valuable information in the "war" against Bin Ladin. The Inquiry's focus on the Intelligence Community limited the extent to which it explored these issues, and this is an area that should be reviewed further.

Discussion: This Inquiry confirmed that, prior to September 11, problems in information sharing reached beyond the boundaries of the Intelligence Community to encumber the flow of information to and from various other entities. At each level, communications with potentially valuable partners in the war against terrorism - other

84

federal agencies, state and local authorities -- were restricted. Witnesses testified that these restrictions on information flow occurred at great cost to the counterterrorism effort.

Officials in the Departments of Treasury, Transportation, and State told the Joint Inquiry that, although they receive threat information from the Intelligence Community, they do not always receive the information that adds context to the threat warnings. In many instances, officials told the Joint Inquiry, this lack of context prevents them from properly estimating the value of the threat information and taking preventive actions. The Joint Inquiry was also told that not all threat information in the possession of the Intelligence Community is shared with non-Intelligence Community entities that need it the most in order to counter the threats.

For example, DCI Tenet testified that: "The documents we've provided show some 12 reports spread over seven years which pertain to possible use of aircraft as terrorist weapons. We disseminated those reports to the appropriate agencies, such as the [page 90] FAA, the Department of Transportation and the FBI as they came in." Subsequently, the Transportation Security Intelligence Service (TSIS) -- which formerly was the Intelligence Office at FAA -- researched the 12 reports mentioned by DCI Tenet to determine what actions had been taken as a result. TSIS reported to the Joint Inquiry that it had no record of having received three of those reports, two others had been derived from State Department cables, and one report was not received at all by FAA until after September 11, 2001. A TSIS official also testified that, despite its clear relevance to civil aviation, the FAA was not provided a copy of the FBI's July 1, 2001 Phoenix communication until its existence was made known to officials there by the Joint Inquiry in early 2002.

In a similar vein, the FAA had certain intelligence information in its possession prior to September 11 regarding the terrorist who was apprehended on his way from Canada to the Los Angeles Airport at the time of the Millennium. It also had conducted a detailed analysis of the bomb materials that were seized with him, and connected them to the Bojinka Plot to blow up commercial airliners over the Pacific that had been discovered in the Philippines in 1995. In testimony to the Joint Inquiry, a TSIS official

85

indicated uncertainty regarding whether or not these findings had been formally communicated to the CIA.

The CIA and NSA had sufficient information available concerning future hijackers al-Mihdhar and al-Hazmi to connect them to Usama Bin Ladin, the East Africa embassy bombings, and the USS Cole attack by late 2000, and there were at least three different occasions when these individuals should have been placed on the State Department's TIPOFF watchlist and the INS and Customs watchlists. Nonetheless, this was not done, nor was the FBI notified of their potential presence in the United States until late August 2001.

The CIA also did not provide the Department of State with almost 1500 terrorism related intelligence reports until shortly after September 11, 2001. These reports led to the addition of almost 60 names of terrorist suspects to the State watchlist. Also, due to a [page 91] lack of awareness of watchlisting policies and procedures among CIA personnel before September 11, this information was not provided to the watchlists at INS, and Customs. Intelligence officers at the Departments of Energy and Transportation also did not have access to FBI data, CIA reports, and names on the watchlists.

The FBI did not advise the Department of State's Diplomatic Security Service of the reasons for its inquiries regarding al-Mihdhar and al-Hazmi's visa information in August 2001 when it was engaged in efforts to find the two individuals in the United States. Neither was INS asked by the FBI to use means available to it, including a search of the Law Enforcement Support Center's database, to locate al-Mihdhar and al-Hazmi when the FBI was looking for them in the United States in August 2001. INS and FAA officials who testified at the Joint Inquiry's October 1, 2002 hearing asserted that their agencies might have been able to assist the FBI in locating the two if the FBI had told them of the purpose and importance of the search.

Officials from the Departments of Transportation, State, Energy, Defense, and Treasury stated to the Joint Inquiry that, unless information is shared by the Intelligence Community on a timely basis, they are unable to include dangerous individuals on various watchlists to either deny them entry into the United States or apprehend them in

86

the United States. The Transportation Security Administration Assistant Under Secretary of Intelligence testified that, had he received the July 2001 FBI Phoenix field office agent's Electronic Communication, for example, he would have "…started to ask a lot more probing questions of the FBI as to what this was all about…what connections these people may have had to flight schools, by going back to the Airmen Registry in Oklahoma City that is maintained by the FAA to try to identify additional people."

The INS also was not privy to the presence of two known terrorists inside the United States. The INS Deputy Executive Associate Commissioner testified to the Joint Inquiry hearing on October 1, 2002 that "there is a likelihood" that INS agents would have been able to stop al-Mihdhar and al- Hazmi in August 2001. The INS Law Enforcement Support Center (LESC) has been in operation for more than ten years. It is [page 92] capable of querying every INS database and is available on a 24 hour per day, seven-day per week basis. The LESC reportedly can provide information in about seven minutes on the legal status of individuals in the United States.

In their testimony before the Joint Inquiry hearing, state and local government witnesses were adamant about the necessity of the intelligence and law enforcement agencies sharing terrorist information with state and local authorities. Former Virginia Governor James Gilmore stated that, in his entire four-year term, he never received any intelligence or law enforcement information regarding terrorists. Governor Gilmore also testified that:

. . . to the extent that there has been intelligence sharing, it has been ad hoc. It has been without a real systematic approach. And what would you expect. With the Intelligence Comnunity, it is not within the culture if not within the statute that you don't share information. If you do, you are even subject to criminal penalties not to mention the danger of sharing information and to the danger of people who provide it. And the capacities of the United States in order to gather it.

In addition, he explained that he was not even given a security clearance while he was Governor that would have allowed him to be briefed on possible terrorist plots.

The Police Commissioner of Baltimore stated at the same hearing that he does not receive intelligence information about suspected terrorists living in his jurisdiction even

87

though some may have been associated with the September 11 hijackers. He also cited the fact that there are 650,000 law enforcement officers nationwide and they should be viewed by the federal intelligence and law enforcement agencies as force-multipliers. However, this can only happen if information flows in both directions. The Police Commissioner also testified that, domestically, the local police force is the "biggest collector" of information, not the federal government. To illustrate his point that information must flow in both directions, he added, "we can tell when people move from one cave to another in Afghanistan, but we can't tell when they move from one row house to another in Baltimore."

By contrast, former FBI Director Louis Freeh testified that information sharing with federal, state and local authorities was a priority for the FBI. In his Joint Inquiry testimony on October 8, 2002, he said: [page 93]

We doubled and tripled the number of Joint Terrorism Task Forces [JTTFs] around the United States so we could multiply our forces and coordinate intelligence and counterterrorism operations with the FBI's federal, state, and local law enforcement partners. Thirty-four of these JTTFs were in operation by 2001. . . . We were also tasked to set up the National Domestic Preparedness Office to counter terrorist threats and to enhance homeland security.

Mr. Freeh added that counterterrorism was such a high priority that the FBI instituted a national threat warning system in order to disseminate terrorism related information to state and local authorities around the country and organized national, regional and local practice exercises to help the country prepare for terrorist attacks.

Further, FBI Director Mueller explained in his October 17, 2002 testimony before the Joint Inquiry the changes that had been made in this regard by the FBI since September 11, and added that:

As a result of these initiatives and despite some of the testimony that this [Inquiry] has heard, we have received numerous letters of support and gratitude from state and local officials and most particularly from the International Association of Chiefs of Police [IACP]. . . . Our agents must work closely with our local and state law enforcement partners. . . . I don't believe that [the testimony of the Baltimore Police Commissioner] is representative of the feeling in the field. Does his testimony surprise me? I would say probably not. But I will tell you every time that I have . . .

88

.seen, either publicly or in testimony before this committee or another committee, that there is a police chief who is not getting what he or she wants, I have called, picked up the phone and called them to try to address those concerns.

 . . .

[A] letter from William Berger, the President of the IACP. . . . praises us for the changes we have made to address this particular problem. I will just read one paragraph:

It is my belief that the steps you have taken have been very responsive to these concerns and clearly demonstrate the FBI's commitment to enhancing its relationship with State and local law enforcement in improving our ability to combat not only terrorism, but all crime.

I was at the IACP two weeks ago. I talked to the hierarchy, and I believe that they are supportive. There are isolated individuals throughout the United States who do not believe we are doing enough, and there are areas where we still have a ways to go, getting clearances for chiefs of police, exchange of information all the [page 94] way down and getting it back up. We have a number of [JTTFs] that are working exceptionally well around the country. I think if you went to 9 or 10, or 99 out of 100, or 55 out of 56 you will find that State and local police are very supportive of the relationship. There will always be one, there will always be two, and we try to address them as we come along.

Following the events of September 11, 2001, the IACP President did indeed write to FBI Director Mueller to express his appreciation for the steps the FBI has taken, including the creation of the State and Local Law Enforcement Advisory Panel and the Office of Law Enforcement Coordination. Subsequently, however, the IACP President was quoted on September 19, 2002 that:

[Federal communications with state and local police] didn't work again…Most local police in New England were informed by the FBI office in that area…about an hour before the public, but police in other regions didn't know about the change until Attorney General John Ashcroft and Homeland Security Director Tom Ridge announced it at a press conference.

The Inquiry found that the FBI's establishment of JTTFs in many FBI field offices had begun to correct some information sharing problems by encouraging coordination between federal, state, and local agencies prior to September 11. These efforts did result in some successes. For example, in the Moussaoui investigation, the INS representative on the Minneapolis JTTF was able to use the INS database to

89

determine Moussaoui's immigration status quickly. The INS and FBI representatives then approached Moussaoui together and he was taken into INS custody at an INS facility and questioned by the FBI.

However, a variety of shortcomings in the JTTF program limited its effectiveness prior to September 11. First, not all of the FBI field office had JTTFs. Further, some of the JTTFs were hampered by a lack of analytic personnel, limited participation by local law enforcement organizations, incomplete access to information by some of the participants, and the absence of CIA detailees.

Prior to September 11, only 35 FBI field offices had JTTFs and only six JTTFs had CIA representatives. This might help explain why [page 95] the CIA did not receive a copy of the July 2001 Phoenix communication until well after September 11. The Gilmore Advisory Panel reported anecdotal evidence suggesting that the JTTF and other similar efforts, while well intentioned, continue to be confusing, duplicative, non-routine, and bifurcated in both structure and implementation.

11. Finding: Prior to September 11, 2001, the Intelligence Community did not effectively develop and use human sources to penetrate the al-Qa'ida inner circle. This lack of reliable and knowledgeable human sources significantly limited the Community's ability to acquire intelligence that could be acted upon before the September 11 attacks. In part, at least, the lack of unilateral (i.e., U.S. -recruited) counterterrorism sources was a product of an excessive reliance on foreign liaison services.

Discussion: The U.S. Intelligence Community was not able to penetrate al-Qa'ida's inner circle successfully before September 11, despite the fact that human penetration of that organization was considered a priority. Richard Clarke, the former National Counterterrorism Coordinator, described the problem as well as the impact that it had on policymakers:

[It was not until 1999 that the Counterterrorism Center began to have some success in developing penetrations of al-Qa'ida. A new Director. . .took over the Counterterrorism Center and was instructed by George Tenet to get human penetrations of al-Qa'ida, and they did have some success in the succeeding years, although none of them very high level.

90

. . . .

. . . . [ ] never had anyone in position to tell us what was going to happen in advance, or even where Bin Ladin was going to be in advance [ ] we never knew where he was going to be in advance, And usually we were only informed about his - where he was after the fact. . . . And [ ] where they were able to tell us where they thought he was at the moment, [ ] the CIA itself recommended against action, because they said their sources were not very good, or not good enough to recommend military action].

[Page 96]

Former Director Louis Freeh emphasized the critical difference that human sources and adequate "infiltration" of terrorist organizations could have made in the context of the September 11 attacks:

If one of those 19 hijackers had spoken - maybe they did, maybe we don't know about it yet - incautiously or imprudently to someone in some place where that information could have been captured, we could have had a day of terror prevented instead of September 11th. There's all kinds of possibilities there. So, infiltration. We need to have our agents sitting around wherever they were sitting around in Hamburg and the U.A.E. and other places, as well as in the caves over in Afghanistan so we can know what is going on.

[Lacking access to senior, high level al-Qa'ida leadership, the Community relied on secondhand, fragmented and often questionable human intelligence information, a great deal of which was obtained from volunteers or sources obtained through the efforts of foreign liaison].

[According to senior CTC officials, CIA had no penetrations of al-Qa'ida's leadership and never obtained intelligence that was sufficient for action against Usama Bin Ladin from anyone. A large number of current and former CTC officers indicated that CTC had numerous unilateral sources outside the leadership who were reporting on al-Qa'ida, and a larger number who were being developed for recruitment, prior to September 11. The best source was handled jointly by CIA and the FBI. In addition, CIA managed a network of [ ] in Afghanistan that often reported information regarding Bin Ladin issues and relations with the Taliban. They occasionally provided threat information as well, but had no access to al-Qa'ida's leadership].

91

[Especially after the East Africa U.S. embassy bombings in 1998, CIA also tried many avenues in an effort to obtain access to Bin Ladin and his inner circle. [ ] [page 97] [ ]. Despite these creative attempts, according to former senior officials of CTC, CIA had no penetrations of al-Qa'ida's leadership, and the Agency never acquired intelligence from anyone that could be acted upon, prior to September 11].

[Numerous sources were being handled by foreign intelligence services. Most disruptions of al- Qa'ida activities abroad before September 11 were the result of joint initiatives with foreign governments. However, relying on foreign services [ ] meant that very little counterterrorism intelligence was obtained by CIA in some parts of the world [ ].

[There was a surge in volunteer sources after the 1998 East African embassy bombings, another surge on the anniversary of those bombings in 1999, and a third after the December 1999 disruption of the Jordanian Millennium plot. [ ]. One of these was very good and provided information that was used to thwart attacks on U.S. interests in Europe. Several of these volunteers continue to act as CIA sources. [ ]. The negative considerations were that most volunteer information was considered bogus by CTC, some volunteers were suspected of being al-Qa'ida provocations, and some were believed to have cooperated with terrorist groups].

92

The inability to develop reliable human sources effectively stemmed, in part, from the difficult nature of the target. Members of Usama Bin Ladin's inner circle have close bonds established by kinship, wartime experience, and long-term association. [Page 99] Information about major terrorist plots was not widely shared within al-Qa'ida, and many of Bin Ladin's closest associates lived in war-torn Afghanistan. The United States had no official presence in that country and did not formally recognize the Taliban regime, which viewed foreigners with suspicion. Pakistan is the principal access point to southern Afghanistan, where al-Qa'ida was particularly active, but U.S.- akistani relations were strained by Pakistani nuclear tests in 1998 and a military coup in 1999.

While attempts to penetrate al-Qa'ida cells outside Afghanistan may have presented fewer obstacles, other factors limited CIA efforts to do so. [ ]. This meant as a practical matter that CIA did not focus as heavily as would otherwise have been the case on recruiting human sources of counterterrorism intelligence in other locations such as [ ].

CTC personnel said they did not view guidelines issued by former DCI John Deutch in 1996 concerning CIA recruitment of human sources with poor human rights records as an impediment to the pursuit of terrorist recruitments in al-Qa'ida, and none of the CTC officers interviewed by the Joint Inquiry attributed the lack of penetration of the al-Qa'ida inner circle to the Deutch guidelines. In fact, the effort to recruit such penetrations became increasingly aggressive with respect to Bin Ladin's network beginning in 1999. These responses should be balanced against the examination of the effect of the Deutch guidelines that was conducted by the House Permanent Select Committee on Intelligence (HPSCI) Subcommittee on Terrorism and Homeland Security. Its July 2002 report stated in this regard: [page 99]

93

Go to Next Page